from einops import rearrange

from diffusers.models.attention_processor import *


# Copied from diffusers.models.attention_processor.JointAttnProcessor2_0
# The only modifications: reshape qkv by `num_views`
class JointMVAttnProcessor2_0:
    """Attention processor used typically in processing the SD3-like self-attention projections."""

    def __init__(self):
        if not hasattr(F, "scaled_dot_product_attention"):
            raise ImportError("AttnProcessor2_0 requires PyTorch 2.0, to use it, please upgrade PyTorch to 2.0.")

    def __call__(
        self,
        attn: Attention,
        hidden_states: torch.FloatTensor,
        encoder_hidden_states: torch.FloatTensor = None,
        attention_mask: Optional[torch.FloatTensor] = None,
        num_views: Optional[int] = None,
        *args,
        **kwargs,
    ) -> torch.FloatTensor:
        residual = hidden_states

        batch_size = hidden_states.shape[0]

        # `sample` projections.
        query = attn.to_q(hidden_states)
        key = attn.to_k(hidden_states)
        value = attn.to_v(hidden_states)

        # Multi-view Self-Attn
        num_views = num_views if num_views is not None else kwargs.get("num_views", 1)
        key = rearrange(key, "(b v) n d -> b (v n) d", v=num_views)
        value = rearrange(value, "(b v) n d -> b (v n) d", v=num_views)
        query = rearrange(query, "(b v) n d -> b (v n) d", v=num_views)
        batch_size //= num_views

        inner_dim = key.shape[-1]
        head_dim = inner_dim // attn.heads

        query = query.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
        key = key.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
        value = value.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)

        if attn.norm_q is not None:
            query = attn.norm_q(query)
        if attn.norm_k is not None:
            key = attn.norm_k(key)

        # `context` projections.
        if encoder_hidden_states is not None:
            encoder_hidden_states_query_proj = attn.add_q_proj(encoder_hidden_states)
            encoder_hidden_states_key_proj = attn.add_k_proj(encoder_hidden_states)
            encoder_hidden_states_value_proj = attn.add_v_proj(encoder_hidden_states)

            encoder_hidden_states_query_proj = encoder_hidden_states_query_proj.view(
                batch_size, -1, attn.heads, head_dim
            ).transpose(1, 2)
            encoder_hidden_states_key_proj = encoder_hidden_states_key_proj.view(
                batch_size, -1, attn.heads, head_dim
            ).transpose(1, 2)
            encoder_hidden_states_value_proj = encoder_hidden_states_value_proj.view(
                batch_size, -1, attn.heads, head_dim
            ).transpose(1, 2)

            if attn.norm_added_q is not None:
                encoder_hidden_states_query_proj = attn.norm_added_q(encoder_hidden_states_query_proj)
            if attn.norm_added_k is not None:
                encoder_hidden_states_key_proj = attn.norm_added_k(encoder_hidden_states_key_proj)

            query = torch.cat([query, encoder_hidden_states_query_proj], dim=2)
            key = torch.cat([key, encoder_hidden_states_key_proj], dim=2)
            value = torch.cat([value, encoder_hidden_states_value_proj], dim=2)

        hidden_states = F.scaled_dot_product_attention(query, key, value, dropout_p=0.0, is_causal=False)
        hidden_states = hidden_states.transpose(1, 2).reshape(batch_size, -1, attn.heads * head_dim)
        hidden_states = hidden_states.to(query.dtype)

        if encoder_hidden_states is not None:
            # Split the attention outputs.
            hidden_states, encoder_hidden_states = (
                hidden_states[:, : residual.shape[1]*num_views],
                hidden_states[:, residual.shape[1]*num_views :],
            )
            if not attn.context_pre_only:
                encoder_hidden_states = attn.to_add_out(encoder_hidden_states)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        # Multi-view Self-Attn
        hidden_states = rearrange(hidden_states, "b (v n) d -> (b v) n d", v=num_views)
        batch_size *= num_views

        if encoder_hidden_states is not None:
            return hidden_states, encoder_hidden_states
        else:
            return hidden_states


# Copied from diffusers.models.attention_processor.AttnProcessor2_0
# The only modifications: reshape qkv by `num_views`
class MVAttnProcessor2_0:
    r"""
    Processor for implementing scaled dot-product attention (enabled by default if you're using PyTorch 2.0).
    """

    def __init__(self):
        if not hasattr(F, "scaled_dot_product_attention"):
            raise ImportError("AttnProcessor2_0 requires PyTorch 2.0, to use it, please upgrade PyTorch to 2.0.")

    def __call__(
        self,
        attn: Attention,
        hidden_states: torch.Tensor,
        encoder_hidden_states: Optional[torch.Tensor] = None,
        attention_mask: Optional[torch.Tensor] = None,
        temb: Optional[torch.Tensor] = None,
        num_views: Optional[int] = None,
        *args,
        **kwargs,
    ) -> torch.Tensor:
        if len(args) > 0 or kwargs.get("scale", None) is not None:
            deprecation_message = "The `scale` argument is deprecated and will be ignored. Please remove it, as passing it will raise an error in the future. `scale` should directly be passed while calling the underlying pipeline component i.e., via `cross_attention_kwargs`."
            deprecate("scale", "1.0.0", deprecation_message)

        residual = hidden_states
        if attn.spatial_norm is not None:
            hidden_states = attn.spatial_norm(hidden_states, temb)

        input_ndim = hidden_states.ndim

        if input_ndim == 4:
            batch_size, channel, height, width = hidden_states.shape
            hidden_states = hidden_states.view(batch_size, channel, height * width).transpose(1, 2)

        batch_size, sequence_length, _ = (
            hidden_states.shape if encoder_hidden_states is None else encoder_hidden_states.shape
        )

        if attention_mask is not None:
            attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length, batch_size)
            # scaled_dot_product_attention expects attention_mask shape to be
            # (batch, heads, source_length, target_length)
            attention_mask = attention_mask.view(batch_size, attn.heads, -1, attention_mask.shape[-1])

        if attn.group_norm is not None:
            hidden_states = attn.group_norm(hidden_states.transpose(1, 2)).transpose(1, 2)

        query = attn.to_q(hidden_states)

        if encoder_hidden_states is None:
            encoder_hidden_states = hidden_states
        elif attn.norm_cross:
            encoder_hidden_states = attn.norm_encoder_hidden_states(encoder_hidden_states)

        key = attn.to_k(encoder_hidden_states)
        value = attn.to_v(encoder_hidden_states)

        inner_dim = key.shape[-1]
        head_dim = inner_dim // attn.heads

        # Multi-view Self-Attn
        num_views = num_views if num_views is not None else kwargs.get("num_views", 1)
        key = rearrange(key, "(b v) n d -> b (v n) d", v=num_views)
        value = rearrange(value, "(b v) n d -> b (v n) d", v=num_views)
        query = rearrange(query, "(b v) n d -> b (v n) d", v=num_views)
        batch_size //= num_views

        query = query.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)

        key = key.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
        value = value.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)

        if attn.norm_q is not None:
            query = attn.norm_q(query)
        if attn.norm_k is not None:
            key = attn.norm_k(key)

        # the output of sdp = (batch, num_heads, seq_len, head_dim)
        # TODO: add support for attn.scale when we move to Torch 2.1
        hidden_states = F.scaled_dot_product_attention(
            query, key, value, attn_mask=attention_mask, dropout_p=0.0, is_causal=False
        )

        hidden_states = hidden_states.transpose(1, 2).reshape(batch_size, -1, attn.heads * head_dim)
        hidden_states = hidden_states.to(query.dtype)

        # linear proj
        hidden_states = attn.to_out[0](hidden_states)
        # dropout
        hidden_states = attn.to_out[1](hidden_states)

        # Multi-view Self-Attn
        hidden_states = rearrange(hidden_states, "b (v n) d -> (b v) n d", v=num_views)
        batch_size *= num_views

        if input_ndim == 4:
            hidden_states = hidden_states.transpose(-1, -2).reshape(batch_size, channel, height, width)

        if attn.residual_connection:
            hidden_states = hidden_states + residual

        hidden_states = hidden_states / attn.rescale_output_factor

        return hidden_states